Instructions for use

Title Comparison of Population Dynamics of Two Pierid Butterflies, Pieris rapae crucivora and P. napi nesis, Living in theSame Area and Feeding on the Same Plant in Sapporo, Northern Japan (With 25 Text-figures and 24 Tables)

Author(s) YAMAMOTO, Michiya

Citation 北海道大學理學部紀要, 22(3), 202-249

Issue Date 1981-02

Doc URL http://hdl.handle.net/2115/27663

Type bulletin (article)

File Information 22(3)_P202-249.pdf

Hokkaido University Collection of Scholarly and Academic Papers : HUSCAP

Comparison of Population Dynamics of Two Pierid Butterflies, Pieris rapae crucivora and P.

napi nesis, Living in the Same Area and Feeding on the Same Plant in

Sapporo, Northern Japan!)

By

Michiya Yamamoto

Zoological Institute, Hokkaido University

(With 25 Text-figure8 and 24 Table8)

The aim of the present study is to compare the population trends of two closely related pierid butterflies, Pieris rapae crucivora Boisduval and P. napi nesis Fruhstorfer, living in the same area and feeding on the same plant. P. rapae crucivora is a well known pest of cultivated crucifers and one of the most predomi­nant butterflies throughout the lowlands of Japan. P. napi nesis feeds almost exclusively on wild crucifers and abundant only in the northern Japan. On the studied area, however, the larvae of both species mainly feed on the exotic weed, Rorippa sylvestris (L.) Besser. Under this rather exceptional situation, P. rapae crucivora is more abundant than P. napi nesis. The elucidation of how the relative abundance of these co-habitants is maintained is the main purpose of this study. Many previous studies on population ecology of insects have dealt with a single pest species. The comparison of two species, one being a pest and the other not, living in the same area involves some interesting aspects so far relatively ignored, e.g. relative abundance and qualification for pest. Main efforts of the study were concentrated to estimate the number of eggs laid per female and to construct the life tables of two species. The former involves also the information on the activity range of adults and the latter the results of key factor analysis.

Materials and methods

1. Distribution of the three pierid butterflies: In Japan there occur three common Pieris species, being closely related each other and all with the Northern

1) This paper constitutes a part of the doctoral thesis presented to the Faculty of Science, Hokkaido University.

Jour. Fac. Sci. Hokkaido Univ. Ser. VI, Zool. 22 (3), 1981 202

Comparison of Population Dynamics of Two Pierid Butterflies 203

Palaearctic elements, P. rapae, P. napi and P. melete.1 ) P. rapae is a Holarctic species of the Eurasian origin and lately has invaded by accidents Australia and New Zealand. The populations of East Asia including Japan are regarded as ssp. crucivora Boisduval. P. napi is a Northern Palaearctic species with its southern limit of distribution in Japan. The populations of Honshu and Hokkaido are respectively classified in ssp. japonica Shirozu and nesis Fruhstorfer. The world distribution of two species mentioned above is shown in Fig. 1. The third species, P. melete M?metries is the southernmost representative, showing a Japan­China-Himalayan distribution. These three species exhibit a complete sexual isolation for one another (Suzuki et al. 1977) and the larvae are segregated by the difference of food plants (Hiura 1973, Ohsaki 1978), though the adults are found

Fig. 1. World distribution of P. rapae and P. napi (cf. Riura 1979).

1) Recently the genus Artogeia has been adopted for these species by some authors (e.g. Artogeia rapae, A. napi and A. melete).

204 M. Yamamoto

Table 1. Rank order of five top butterfly species at four localities in Japan. Jugoshima (JM) , western Hokkaido (Yamamoto 1977); Utsunomiya, eastern Honshu (Gilda 1969); Kyoto, western Honshu (Morisita 1967); Hirao, southern Japan (Kawabata 1976). Gradient: Relative dominance according to latitudinal change.

JM I Utsunomiya I Kyoto I Hirao I Graident

Colias erate

I

1 4 15 11 N>S Pieris rapae 2 1 2 1 N=S P. napi

I

3 46 - - N~S Lycaena phlaeas 4 7 12 21 I N>S Everes argiades 5 5 4 7 N=S Y pthima argus

I 8 2 1 2 N<S

Eurema hecabe - 3 3 8 N<S N eope goschkevitschii 17 28 5 9 ? Zizeeria maha I - 11 7 3 N<S

I

I

Papilio xuthus - 19 21 4 i

? Parnara guttata - 12

I 6 5 I N<S

Table 2. Predominant butterfly species in and near Sapporo expressed by Ps (species prosperity index, Morisita 1967) and N (total number of indivudals observed in 1973), Jo (Jozankei), JM (Jugoshima) and UC (the campus of Hokkaido Uni-versity). H: Habitat type preferred (O=open-land, W=woodland, Ow, Ow, oW =intermeidates).

Jo (28.6=mean frequency) I

JM (46.1) I

UC (91.0)

Species Ps (N) HI Species Ps (N) HI Species Ps (N)

Pieris napi 1. 046 (479.0) OW Colias erate 1. 762 (665.5) 0 Pieris 3.132 (1,229.5) P. rapae 0.674 (301. 5) 0 Pieris 1. 621 (503.0) 0 rapae Colia8 erate 0.375 (161.0) 0 rapae CoUas 2.180 778.5) Araschnia 0.104 ( 60.5) W P. napi 1. 051 (384.0) OW erate

H

0

0

burejana Lycaena 0.255 (112.5) 0 Pieri8 0.550 231.5) OW Brenthis 0.087 ( 58.5) W phlaeas napi

ino Everes 0.202 (193.0) OW Lycaena 0.239 132.5) 0 Lycaena 0.079 ( 55.0) 0 argiades phlaeas

phlaeas Ochlodes 0.095 ( 89.0) Ow Evere8 0.142 112.0) OW Papilio 0.070 ( 54.5) Ow venata argiades

machaon Ochlode8 0.066 94.0) OW Aglais 0.076 ( 53.5) W venata

urticae (38.0) OW I Everes 0.040

argiade8 Inachis io 0.081 ( 34.5) oW Thoressa 0.026 ( 34.5) W

varia Argyronome 0.050 ( 29.5) Ow

ruslana Ochlodes 0.035 ( 27.5) Ow

venata ----

Total 1,387.5 1,947.0 2,578.0 % 79.5 84.5 96.5

Comparison of Population Dynamics of Two Pierid Butterflies 205

interminglingly on flight. P. rapae crucivora is abundant throughout Japan and recently introduced in Okinawa (Osada and Ito 1974). This species occupies the top or second rank in many local butterfly assemblages (Table 1). P. napi and P. melete behave oppositely in the latitudinal relative abundance, the former more abundant in the northern areas and the latter in the southern areas. In the areas in and near Sapporo the two pierid butterflies P. rapae crucivora and P. napi nesis, are predominant species together with Oolias erate Esper (Table 2), both being triboltine with a similar phenology. Adults of P. rapae crucivora prefer openlands and insolated areas. P. napi nesis is eurytopic but prefers rather shaded areas, though more abundant in woodlands (Yamamoto 1977). The two forms are henceforth abbreviated as Prc and Pnn.

2. Oensus procedures: Oensus methods adopted are tabulated below, to­gether with the years in which the respective survey was carried out:

Survey for: Adult number

Number of eggs laid by a single female Census of immature stage Number of pupa and adult emergence

Method Belt transect census Collecting for 0.5 hours Capture-recapture method Adult chase Quadrat method Counting along the walls

of buildings

Year 1972, '73 '74 1977, '78 1974 1975, '76 1974 1974 '79

Further details about the procedures are explained in the sections concerned.

3. Area studied: The campus of Hokkaido University (UO) used for the studies (Fig. 2) consists of three different habitats for the two species.

Crop field of wheat, potato, beet, corn and bean.··········· . Areas along B~ D and a part of areas along A~ B in the map.

Abandoned areas with weeds.·· ................................ Areas along A~ Band

Patches of wood, the undergrowth of which sparsely involves Cardamine leucantha O.E. Schulz, an original larval food

E~F.

plant of Pnn.·· ................................................ Areas along D-E.

Oensus belt for adults was about 4.6 km running from A through B, 0 and D to F (Fig. 2, arrow). In total 34 quadrats were set on the area to study the floral composition (Fig. 2, open circles). About 15 of these quadrats (Fig. 2, double circles) in A",O were also used to construct the life tables of immature stage. Several stations for caputre-recapture method were set in A to O. Marked butterflies were released at a single spot (Fig. 2, cross) in the evening. Adult chase was carried out throughout the area of UO to clarify the number of eggs laid by a single female and activity range of adult butterflies.

206 M. Yamamoto

Fig. 2. Map of the campus of Hokkaido University. Open circles: Quadrats used for study of floral composition. Double circles: Quadrats used for the studies of both floral composition and immature life tables. Cross: Releasing point for capture-recapture method.

= Patche. of wood

$S Orchard grass

100m ,

4. Larval food plants: As given introductively, Pre mainly feeds on culti­vated crucifers, while Pnn on wild crucifers. But in western Hokkaido the clear-cut segregation of food plants between two species is obscure due to the existence of a common food plant, Rorippa sylvestris. In and near Sapporo, the main food plants of Pre are Brassiea oleraeea var. eapitat, Raphanus sativus and Rorippa sylvestris, and of Pnn Cardamine leucantha and Rorippa sylvestris. Especially in the area surveyed, crop crucifers have not been abundantly cultivated and the wild crucifers mentioned above were scarce except Rorippa sylvestris. Consequently both species are now highly dependent upon the exotic mustard, Rorippa sylvestris. This weed was first found and recorded by Okuyama (1963) at Ofune in Tokyo. In Hokkaido Sakagami and Fukuda (1973) incidentally reported in their wild bee survey at DC in 1959 the weed as Rorippa sp. in the list of plants predominantly visited by bees. Hasegawa (1966) reported the weed as a common larval food plant between the two pierid species. Later it has spread every­where in and near Sapporo and is extending its range to northern Hokkaido, though less common in Honshu and eastern Hokkaido. Coexistence of the two pierid species found in and near Sapporo had not appeared until this weed species grew abundantly.

Comparison of Population Dynamics of Two Pierid Butterflies 207

Results and discussions

Adult number of a given generation, n, produced in the area is given by the following equation:

in which Na=adult number, Nt=female number, Ne=number of eggs laid by a single female within a limited area, Se=survival rate of egg, Sl=survival rate of larva, Sp=survival rate of pupa, Sa =successful rate of emergence and n=generation number. To clarify factors causing the fluctuation of adult number in DC, exact estimation of all these items is necessary. Ne is obtained by combining egg number per female in her life span with her staying duration (or survival duration) within DC. Obviously the area may receive some immigrating adults while lose some adults by emigration. Immigration is negligible in the present study because of less habitats for two species in the surrounding areas out of DC, whereas emgiration is intense.

The results are divided into four parts, number of adults (for estimation of Na and Nt), number of eggs per female (for estimation of Ne used in combination with the next item), activity range of adults, and life tables of two species (for estimation of Se, SI> Sp and Se). The related discussions are given in each section.

1. The number of adults

The number of adults was estimated by combining the caputre-recapture method and belt transect census, both bearing some merits and demerits. The capture-recapture method is theoretically most relil!-ble but in practice its efficiency is often low in comparison with the effort expended. On the other hand, the belt transect census is rather easily practicable, but it is not easy to estimate the absolute density from the obtained relative abundance. Further, both methods should be repeated to cover the variation range caused by diurnal, seasonal and annual change of the number (Frazer 1973, Yamamoto 1975). In the present study, these defects were partially, even though not completely, removed by the combined use of two complementary methods.

1.1. Comparison between belt transect census method and capture-recapture method: Before dealing with the individual number obtained as relative abundance, the discrepancy between the real and the observed individual number must be discussed based upon the result obtained by the capture-recapture method, which was carried out in parallel with the belt transect method. In adult census made in 1974 any individuals of two species on the wing or at rest within 10 m width of both sides of the census course were registered without collecting. Certain individuals difficult to specify at distance were captured and liberated after identifying species and sex. In the capture-recapture method adults marked with lacquer paint were collected and registered along the census route, together

208 M. Yamamoto

Table 3. Adult number obtained by capture-recapture method in Gh of P. rapae crucivora. Other explanations in text.

Date 11, Ri mi mhi

V 17 31 26 - h=1 -

18 47 45 7 7 2 f---

19 44 42 5 1 4

3 I---

21 45 40 9 2 1 6 4 I---

22 41 38 6 0 2 0 4

Total 208 191 27 10 7 6 4

r,

VI 12 14 13 h=1 r--

13 29 27 3 3 2 I---

14 27 23 8 0 8

3 r---

16 23 17 11 0 4 7

Total 93 80 22 3 12 7

r,

with unmarked ones. Release was made at Point X (Fig. 2) in the evening in order to secure natural dispersal of marked adults. Though releasing was made from a single point, random mingling of marked individuals with unmarked ones seemed to be assured due to high dispersal ability of these species. The results obtained by capture-recapture method are shown in Tables 3",5 on Pre and 6",8 on Pnn. Symbols adopted from Iwao (1971) are: ni=number of individuals sampled at the time i, R;=number of individuals marked and released at the time i, m;= number of marked individuals in ni, mhi=number of individuals recaptured at the time i after last marked at the time hand ri=total number of individuals recaptured in Ri. The result from adult chase described later (Section 3) suggests that DC-population is thought to be open, i.e. allowing dilution and disappearance of some adults. The number of individuals in open population can be estimated by Jolly-Seber's method given by the following equation:

n+l ( R+l ) N i = -'- -'-Zi+mi mi+1 ri+ 1

in which Ni=total number of individuals at the time i, z;=number of individuals marked before i and thereafter recaptured at the time i. Results of calculation

Date

VI 24

27

29

30

VII 1

2

3

4

12

13

14

15

17

18

19

20

Total

Comparison of Population Dynamics of Two Pierid Butterflies

Table 4. Adult number obtained by capture-recapture method in G1

of P. rapae crucivora.

ni Ri mi mhi

15 10 -

h=l t--

26 25 2 2

2 r--

64 60 11 4 7 3 ~-

58 57 23 2 3 18 4 I-

62 62 32 1 1 12 18 5 I-

62 61 29 5 5 19 6 f--

61 60 26 3 6 17 7 t--

57 57 31 1 1 3 10 16

8 t----

79 78 6 2 2 2 9 r--

81 72 17 1 1 15 10 I-

74 71 26 1 12 13 11 t--

38 33 14 1 2 1 10

12 r--

44 40 8 2 3 3 13 t--

59 55 12 1 4 1 6

82 79 23 3 2 5 2

89 84 23 2 1 1 2

951 904 283 9 11 36 27 29 29 19 4 32 21 23 7 8

209

14 I-

11 15 t--

6 11

17 11

~---------------------~ --------------------~

except the data indicating a lower recaptured number (mi~4) were plotted against the number of adults censused on the same day (Fig. 3). The correlation coefficient value is 0.798>0.497 (=95% fiducial limit) in Prc and 0.785>0.514 in Pnn. Although a linear relation is allowed in both cases, an exponential expres­sion using log. estimted number seems more suitable (r=0.81O in Prc and 0.846 in Pnn) by underestimation of too many adults over counting ability. Anyhow, the occurrence of significant correlation between observed and estimated number allows to use the former to analyze fluctuation within the same species. But as fluctuation of individual number varies between two species, a proper correction must be made on account of the different coefficient between two regression functions (Fig. 4):

210

Pre Pnn

M. Yamamoto

log Y=O.00513 X+1.98 log Y=O.01l2 X+1.58

in which Y =number of individuals estimated by capture-recapture method and X= number of individuals censused. The real abundance of two species reverses at the

600 • /

• • / /

/ "C 0- P. ropae crucivora /

,E I

.---- P. nap; nesis /

E ./ • /

~ / ,.

~ 400 / /

e y.5.09X-48;5 0 0

I e /

" / 0 "- /0 ~ / }; 0

/ y-2.5Ix+57.6 / -g • / "5 / • 0

E 200 / 00

~ / 0 0

~ 0 /

" 0 a' 0 o· • " -g I • 15

,. • /

:;; / .c /

is • /

Z

0 50 100

Number of adults observed

Fig. 3. The relation between the observed and the estimated number of adults.

3

•

•

o

50

" y·0.OIl2x+I.58 /,,' . . .......... . ~" ,.,"

•

o ........ 0

//0 Y'0.00513x+I.98 00

•

0- P. rapae crucivora

• - --- P. nap! nosis

100

Number of adults observed

Fig. 4. The relation between the number of adults observed and the logarithmic number of adults estimated based upon capture-recpature method.

Date

VIII 5

9

10

11

12

20

21

22

27

28

30

31

IX 4

5

11

12

13

Total

Oomparison of Population Dynamic8 of Two Pierid Butterflies

Table 5. Adult number obtained by capture-recapture method in G2

of P. rapae crucivora.

ni Ri m, mhi

69 67 -

h=1 I--

78 68 13 13 2 t-----

67 65 20 4 16 3 -

63 60 33 5 12 16 4 '----

66 61 26 2 5 5 14

5 r--

53 47 3 3 6 r--

55 53 11 11 7 t-----

30 28 14 7 7

8 I--

57 55 8 2 3 3 9 r--

51 49 4 1 2 1

10 /--

37 37 3 2 1 11 /--

43 36 4 1 1 1 1

12 r---

25 22 1 1 13 t-----

31 27 6 1 5

14 r--

13 12 0

13 10 2 1

13 11 1 1

764 708 149 24 33 21 14 3 20 13 5 4 2 1 1 7 0

211

15 I--

1 16 -

1 0

~---------- ri ----------~

point of 70 individuals observed. These different coefficients may be caused by a different ratio of active individuals counted at 10 :30-census to total number occur­ring on the same day.

Duration of stay of particular adult (DBA) within studied area with 95% fiducial limit was obtained from the mean interval between marking and the last

212 M. Yamamoto

Table 6. Adult number obtained by capture-recapture method in Gh of P. napi nesis.

Date ni Ri mi mhi

V 17 37 31 - h=l r-----

18 59 58 3 3 2 r-------

19 45 43 1 1

3 t---

21 23 16 2 1 1 4 -

23 13 13 1 1

Total 177 161 7 4 1 1 1

~---- r, -----

Table 7. Adult number obtained by capture-recapture method in G1 of P. napi nesis.

Date ni Ri mi 1nhi

VI 29 11 9 - h=l f--

30 20 19 1 1 2 t---

VII 1 35 35 7 7 3 !---

2 55 53 10 1 1 8 4 t---

3 60 59 17 1 3 13 5 t---

4 58 58 9 2 7

6 !---

12 93 81 2 1 1 7 !---

13 102 79 11 1 10 8 f--

14 87 77 17 7 10 9 f--

15 62 60 15 2 4 9

10 -

17 43 35 18 4 3 4 7 11 -

18 37 32 8 3 3 2 12 t---

19 34 31 12 1 1 1 2 7 13 f--

20 42 33 8 1 1 2 4

Total 739 661 135 2 9 11 17 7 1 24 19 17 10 5 9 4

~----------- r, ---------~

Date n,

Vlll 5 14

9 50

10 59

11 84

12 69

20 116

21 71

22 42

27 63

28 52

30 47

31 63

IX 4 14

5 25

Total 769

Comparison of Population Dynamics of Two Pierid Butterflies

Table 8. Adult number obtained by capture-recapture method in G. of P. napi nesis.

R, In, mhi

12 -

h=1 f--

44 1 1 2 t---

56 4 4 3 -

82 11 1 3 7 4 f--

66 18 4 4 10

5 r--

107 1 1 6 f--

65 14 1 13 7 t---

37 5 2 3

8 -

57 0 9 -

42 8 3 1 4

10 -

39 8 1 7 11 r--

52 8 1 1 2 4

12 t---

10 1 1

19 4 2

688 83 2 12 11 10 1 18 3 2 6 9 4 3

13 -

2

2

~----------- ri ----------

213

recapture date for each generation, post-hibernating generation (Gh), the 1st generation (G1) and the 2nd generation (G2):

Prc Pnn Total Female Total Female

Gh 1.93±0.29(44*) 3.00±2.43(3) 1.71±1.03( 7) 3.00 ( 2) 3.92±2.21(14) 4.00±4.50( 4)

G1 3.36±0.43(177) 2.75±1.60(8) 2.53±0.45(92) G. 3.57±0.52(192) 2.00 (2) 3.14±1.01(65)

*=number of individuals recaptured

These figures are used to avoid double counts of the same individuals on estimating total adults emerged on DC.

214 M. Yamamoto

Table 9. Diurnal change of the number of adults observed.

, Pre I Pnn

D.:imoi-6-7_:3_0_;--i-6-10~~:::--?-i 61_4_:3_:_-_~_., I ,7':-__ 1_!:~ I ,H~-_ V ~ I 3 I 1; ~ i'" 5 8 4

1 15 23 18 5 7

2 4 6 2 12 12

19 26

VI 3 6

11 22 26

VII 1 6

12 16 21 26

VIII 6 11 21 26

IX 1 6

13 21 i

23 21 1 12 8 5 10 6 3

3 3 23 8 5 86 36 10

230 114 19 235 94 34 115 45 12

74 25 6

15 8 12 1 6 1 5 1 6 1

7 2

25 21 5 41 19 16 41 19 15 25 15 7 18 10 5 15 6 3 24 6 7 77 9 10

147 29 14 193 20 23 148 21 23 180 31 18 143 14 15 71 4 4

139 39 131 98 22 40 8 34 3 15 6 3 13 13 1 22 13 3

30 14 2 2 2

23 7 7 20 1 3

7 45

124 174 105

53 24 90

10

3 4 3 8 8 8 14 4 13

1 9

9 2 14 2 20 3 1

3 1 2 2

1 7

34 11 4 6 3 8

3

12 7 6 6 6 4 1 2

o 1 1

2

15 4 1 42 17 4

9 4 5 4 4 4 2 2 1

1 1 6 4

23 19 22 15 4 5 9 9 5 9 3 3

30 14 18 14 4 4 7 4 4 3 7 4

2

1

o 1 1

3 17 31 11 3 5 6

1 5 2

14 11 15 4 7 2

4 3 1 1 9 8

5

2 4 6 3 1 1

o

o

2

1.2.. Diurnal change of the number of adults observed: Diurnal change of the number of individuals observed was studied by belt transect census made three times per day (early morning 7 :30"" late morning 10 :30"" early afternoon 14:30"" each time about 1.5 hrs.) at A", F, UC in the active season of 1973 (Table 9). Both species are very similar in diurnal activity and phenology. In spring (Gh ) a high activity is exhibited in late morning to early afternoon, especially early May (Pnn) and late May (Pre). In both species G1 in early summer indicates a high activity in the early and late morning. Especially Pre females were abundant in the first census. The diurnal activity of G2 in late summer resembled that of Gh , that is, a high activity was shown in the 2nd and 3rd censuses. This phenological change of the diurnal activity in Pre was reported by Hirose (1954), too. To standardize the difference of relative abundance among seasons and between species, census time with most abundant individuals observed should be adopted. In the present case and 2nd (10 :30) census is used as representative for the relative abundance.

1.3. Seasonal change of the number of adults observed: Fig. 5C shows the number of individuals obRerved by adult census made from 10:30 everyday in the

CompariBon of Population Dynamics of Two Pierd Butterflies 215

flight season of two species, 1974. Gh adults appear from middle April to late June, G1 from early July to early August and G2 from early August to early September. Pupae of Gs hibernate and emerge in the next spring, i.e. Gh =Ga. Phenology of

150

100

50

0

50

100

"0 Q)

200 c: Q) <J)

.c 150 0

<J)

C 100

'" "0 .;; 50

'6 .!:

'0 0

~

'" .c E '" z

A 1972

, .. \i\ '{

B 1973 CJ Male • Female

P. rapae crucivoro

Apr May Jun Jul Aug Sep Oct

Fig. 5. Seasonal fluctuation of the adult number of species censused in 1972 ~ 1974.

two species is very similar, both characterized by trivoltinism, pupal hibernation and male predominance. Seasonal fluctuation of the adult number mainly depends upon that of males. Cumulative number of adults counted at 10 :30-census is shown for each generation as follows:

Pre Pnn Total Females Total Females

Gh 758 (49 days) 260 (37 d.) 451 (33 d.) 175 (29 d.) G1 2,662 (33 d.) 457 (42 d.) 1,738 (39 d.) 663 (36 d.) G2 1,718 (43 d.) 366 (40 d.) 2,269 (43 d.) 840 (41 d.)

Considering the different DSA within the census area between species and among generations (p. 213), estimated total number of adults in one generation is given by

216 M. Yamamoto

accumulation of estimated number during a generation divided by the average DSA of individuals for each generation concerned.

In Pre

In Pnn

Total Gh 2,911 (2,530~3,425) G1 2,539 (2,251~2,912) G2 1,844 (1,610~2,159)

Gh 1,042 ( 650~2,621)

G1 1,834 (1,557~2,231) G2 2,030 (l,533~2,988)

Females 1,280 ( 707-6,673) 1,659 (1,049-3,966)

425 560 (358~1,284) 661 (309~?)

In Gh and Gl population a higher abundance of Pre than Pnn is indicated in both males and females. Pre in Gl was numerically superior to that of Pnn, though a higher and acuter increase of the latter species is seen in early July (Fig. 50). In G2 Pnn showed a higher estimated number than Pre, but the relation is obscure in females due to inaccurate estimation of the average DSA. Totally the relation of abundance between these two species and among generations is shown as follows: Pre; Gh >Gl >G2• Pnn; G2 >Gl >Gh • Pre,>Pnn in spring (Gh ), Pre>Pnn in early summer (Gl ) and Pre<Pnn in late summer (G2). Whether this relation is annually maintained or not is discussed in the following section.

1.4. Annual change of the adult number in two species: Unfortunately the an­nual fluctuation of adult number was not continuously traced beyond three years. The comparison of Figs. 5 and 6, both given by unconnected number, suggests a gradual drop of the predominance of Pre from 1972 to 1974 possibly followed by a stable state until 1979, though seasonal pattern of fluctuation of individual

2.0 - - - - - -

~ ~

" '0 E ~

" 1.0 t=

\f'f °567895678956789

1972 1973 1974 1977 1978 1979

Fig. 6. The annual fluctuation of the female ratio between two species (PnnfPrc).

Oompairson of Population Dynamics of Two Pierid Butterflies 217

number seemed to change annually. For Pre the post-hibernating pupae were collected in April from 1975 to 1979 (for 1974, in the preceeding November) and the emergence of adults was checked (Table 10). The result also seems to suggest a gradual decrease of Pre adults, mentioned above.

Table 10. The number of post-hibernating pupae of P. rapae crucivora collected in April (in the preceeding November in 1974).

The number of emergence given in parentheses.

Section yearl 1974 '75 '76 '77 '78 '79 Total

Aa 0 3(3) 3(3) 1 1 (1) 0 8(7) Ab 1 17 (6) 19 (9) 8 (6) 21 (6) I 14(3) 79 (30) Ac 3(3) 0 0 4(3) 2 1 7(3) Bd 1 7(2) 4 (2) 0 0 3 14(4) Be 1 (1) 4 1 2 1 (1) 1 (1) 9 (2)

Total 6 (4) 31 (11) 27 (14) 15 (9) 25(8) 19 (4)

2. The number of e~~s laid per female

It is diffic ult to estimate the fecundity of insects which, as in many butterflies, lay eggs one by one, not in batch between flights (Kurata 1955, Iwata 1967). Even for such the commonest species as P. rapae the estimated number of eggs laid per female differs among authors: 350, Richards 1940; 400, Iwata 1967; 200",300, Ito et al. 1975. These authors do not mention how the fecundity was determined. Recently the fecundity of Pre higher than so far reported was shown under both laboratory (Suzuki 1978) and natural conditions (Yamamoto and Ohtani 1979). These two results were very similar each other in both the number of eggs laid and the daily laying pattern. The number of eggs laid per female of Pnn except Gh was reported by Yamamoto (1978). In the present section the number of eggs laid per Gh-female and the difference among some localities are discussed based upon the dissection of overies.

2.1. The number of eggs laid by a Gh1emale: A total of 35 Pre females and 28 Pnn females collected on UC from May to late June, 1978 were dissected. The results obtained are shown in Table 11, together with the results of G1 and G2 in the same year for each age class classified by condition of scale detachment. The relative age was distinguished into four classes by decrease of the areas covered with scales on abdominal dorsum (I. zero, the dorsum completely covered with scales, II. more than a half covered, III. less than a half covered, IV. nearly exposed). The number of eggs laid is approximately estimated by the discrepancy of the average oocyte number between newly emerged females and old exhausted ones. The results obtained show a similar number of eggs laid by Gh- and G1- female in respective species, though the average female size is different among genera­tions. The number of eggs laid per female of UC-population is summarized,

218 M. Yamamoto

Table 11. Number of oocytes in dissected ovaries with the number of mature eggs in parentheses and S.D. at VC in 1978.

Age class

Date

PreV 1-10 1

11-20, 21-31

VI 1-10 11-20 21-30

VI 21-30 VII 1-10

11-20 21-31

VIII 1-10 11-20 21-31

Pnn V 11-20

21-31 VI 1-10

VII 1-10 11-20 21-31

VIII 1-10 II

11-20 21-31 'I

IX 1-10 ----" ..

I

792( 24)

822 ( 46) ± 16

814 ( 70)

I,064( 96) 847 (104) ±119 796 ( 12)

667 ( 51) ± 42 437 ( 53) ± 18

523( 59)

480( 55)± 88

475( 37) ± 34 434( 66) ± 55

" .0

~ Z

150

II

695 (123) 900 (126) ± 53 674 (114)

938 (154) ±235 1,122 (194)

850 (169) ± 78 615 ( 71) ± 125 511 ( 51) ± 71 560 ( 75) ± 96

477 ( 72) ± 53 449 ( 76) ±227

525 ( 85) ± 59 369 ( 41) ±137 386 ( 98) 436 ( 73) ±139 353( 60)±117 379( 67) ± 92 172( 46)

III

709( 89)±114 701 (114) ±123 657(103) ± 47 513 ( 89)

298 ( 58)

654 ( 70) 998(174) 451( 81) ±125 456 ( 72) ±221 489 (129)

500( 77) ± 57 490( 86)± 45 378( 78)± 68

569 ( 97) ± 64 494 ( 54) ± 116 319( 59)± 41 372 ( 68) 269( 37) ± 98 309( 53) ± 10 273( 37) ± 49

Age In days since emergence

IV

812 ( 76) ± 90 616 (133) ± 40

107 ( 27) 144 ( 16)

158 ( 54)

472( 69) ± 41 380( 92) ± 84 268( 28)

233( 47) ±178 159( 29) ± 43

449 ( 46)

172 ( 28)

Fig. 7. Presumable laying curves of P. rapae erueivora and P. napi nesis and age sequence.

together with the result reported previously (Yamamoto and Ohtani op. cit.): Gh G1 G2

Pre 600-800 600-800 300-700 Pnn 2 0-400 200-400 200-400

Oomparison of Population Dynamics of Two Pierid Butterflies 219

Pre has a fecundity about three times as high as Pnn. Further the laying curve of Pre has a high peak soon after copulation. The difference is expressed as two presumable curve given in Fig. 7 (cited from Yamamoto and Ohtani op. cit.).

2.2. The difference of egg number among localities: To know the local difference of the number of eggs laid, Pre females collected in three districts of southern Japan, Fukuoka City, Yamaguchi City and Osaka City in early spring, 1978, were dissected.

Locality Date No. females dissected Age class II

III IV

Fukuoka IV 4 12

884(62)± 167** 371(43)

*=number of eggs matured **=S.D.

Yamaguchi IV 5 12

1,128(48*) 774(48)± 71 784(56)±l03

Osaka IV 10

8

771(95)± 87 678(38) ± 115

Table 12. Number of oocytes in ovaries of P. napi nesis collected near Sapporo.

Age class

Date I II III IV

Jozankei V 11-20 634 ( 90) - - -

21-31 - 534( 56) ± 12 443 ( 59) -VI 1-10 - - 423( 46) ±144 331 (41) ±126

11-20 546 ( 66) 390( 62) ± 20 - 276 (76) 21-30 - - 332 ( 44) 236 (20)

VII 1-10 - 606 (126)

~~:::~ i - 401 ( 73) ± 113 455 ( 89) ±162 -

474 (106) 368( 71)± 73 214( 66) ± 95 134(16) ± 79 VIII 1-10 164 ( 20) - 205 ( 29) ± 52 122(32) ± 33

11-20 - 309 ( 37) ± 67 242 ( 23) ± 96 150 (27) ± 89 21-31 - 426 ( 34) ± 59 234 ( 38) ± 128 154(29) ± 63

IX 1-10 - - 318( 46) ± 82 154 (17) ± 72 11-20

1

- - I - 236 (36) ±110 I

Hakkenzan V 11-20 504( 0) 425 ( 13) ± 67 349 (31) ± 63 -

21-31 - - 491 (63) ± 10 -VI 1-10 - - 472(58) ± 84 247 (63) ±111

21-30 596 ( 4) - - 340 (40) 68 VII 1-10 510 ( 38) ±116 533 ( 80) ±132 564 (44) 336 (64) ±130

11-20 354 ( 50) 371 ( 51) ± 12 258 (50) 184 ( 0) 21-31 -

I 442 ( 10) 184 (29) ± 18 174 (23) ±1l9

VIII 1-10 - I - 146 (10) -11-20 -

I 184 ( 52) 203 (19) ±106 117 (13) ± 27

21-31 - - 344(59) ±102 195 (31) ± 66 IX 1-10

I

227 (19) ± 10 -I

- -

220 M. Yamamoto

Although most of these females were assumed to be relatively old, the number of oocytes was of the same order to that found in young DC females. This suggests a higher biotic potential of the females of populations in southern Japan.

In Pnn the dissection of ovaries was made with the populations collected at Jozankei (96 females) and Hakkenzan (65) near Sapporo where the larvae feed on Oardamine leueantha (Table 12). The decrease of the average oocyte number between newly emerged females and old exhausted ones was similar to the results obtained on DC (Section 2.1.), but these mountains populations may be bivoltine because newly emerged females were not found in late August when DC-popula­tion issues G2 adult.

The different density of eggs laid in a particular area is caused not only by the different number of eggs laid per female during the life span but also by the different DSA within the area, dispersal ability of females, and number of laying females within the area. This problem is discussed in the next section.

3. Flight activity range of adults

Adult chase or the continuous chase of the particular individual was made in 1975 and 1976 also to know the activity range in adult stage of two species. A total of 15 males and 26 females in Pre (Yamamoto unpub!.), and one male and six females in Pnn were marked soon after emergence or copulation and followed day to day till their disappearance, everyday continuously from start of activities in the morning to roosting in the evening as far as possible. For Pre Ohsaki (1978) reported sedentariness in males and nomadic tendency in females. These tendencies were supported by Kobayashi and Takano (1978), too. As for Pnn no study has hitherto been published on this problem. Fig. 8 shows the regressive oviposition of Pnn-G1-No. ~9 (June 13, 9:00",23, 13:53). The individual made a slight

Fig. 8. The flight activity of P. napi nesis G1 99. Closed circles: Roos­ting sites on successive observation days as given by numerals. Sand E: Start and the end of observations.

Comparison of Population Dynamics of Two Pierid Butterflies 221

emigration from emergence site, then regressively oviposited there. Although both Pre and Pnn have a high flight activity, especially in males

(Ohtani and Yamamoto unpubl.), actual dispersal distance is not so large due to the regressive tendency. This tendency may cause sedentariness of Pre males and P. napi japoniea as reported by Ohsaki (op. cit.)

3.1. The sedentary tendency: The sedentary tendency can be expressed by the duration of stay within a particular area, DSA, a higher value of which indicates a higher sedentariness. For this purpose each individual record was divided by nine concentric circles, Cn(n=l", 9), the center of which is the release point and the diameter is n X 100 m each. By this way the daily activity pattern of any individual can be expressed numerically irrespective of actual course of movement. DSA of an individual within each concentric circle was defined as the period till the first day of emigration with and/or without return from the circle concerned. Fig. 9 shows the average DSA within each concentric circle of activity range in all individuals chased. DSA was longer in males than in females in Pre. Between two species, Pnn was more sedentary than Pre in Grfemales .

Ul

'"' a "0

15

c 10

Ul Q)

E :;:

Q) <.) c Q) "0 'in Q)

cr 5

...... x----x----x----:< //

..... x--- -x- ---x/

- Gh

-GI --e--e--e--e

--- G2

x Mole

o P. rapae crucivora Female

• P. nap; nesis ><---x x--x--x--x--x

~----:- --:=z=: 77-- -rg=g=g=g g ~ --e--e

OL---.---.----r---r--~--_.----r---._--, 2 3 4 5

Diameter

6 7 8 9 XIOOm

Fig. 9. Mean DSA in days within each area (em n=I~9).

3.2. Oomparison in females: In spring both species have a similar DSA with a limited area, while in G1 a higher sedentariness of Pnn is recognized. Although no Pnn-G2-females were chased, their DSA is probably equal or higher than Gl ,

222 M. Yamamoto

judging from the result of capture-recapture method (cf. Section 1.1.). A higher sedentariness of Pnn-females mentioned above affects the estimation of the number of eggs laid within a limited area. Probably the large difference between two species on the number of eggs laid per female during the life span would not be proved if studied within a limited area for a higher sedentariness of Pnn-females. Based upon the presumable curves of laying pattern (Fig. 7), the cumulative number of eggs by a single female is calculated as follows:

Age in days Pre Pnn

I 2 3 4 5 6 7 8 9 10 II 12 13 o 20 80 205 325 415 475 525 565 600 630 660 o 0 0 0 30 60 90 120 150 180 210 240

Within the studied area belonging to the circular area of 900 m in diameter, Pre may lay 200",300 eggs and Pnn about 30 eggs during the first 4",5 days of their DSA in Gh . But in G1 the sedentariness of Pnn-female is three times higher than in Pre. This decreases the interspecific diffrrence and Pnn-female may lay 240 eggs for 12 days in the circular area of 500",900 m in diameter, while Pre--female the same number of eggs in Gh , 200~ 300 for 4",5 days in the same area. In G2

females of Pre stay there for 6", 7 days but lay only 300",400 eggs due to the lower fecundity per female than in Gh and G1. The number of eggs laid by a female within the studied area of UC is thus assumed as follows:

Pre Pnn

Gl

200-300 200-400

G2

300-400 200-400

4. The life tables of two species

The last problem concerns preparation of the life table for each generations, decrease of the number of individuals through successive developmental stages; egg, 1st to 5th instar (L1 ", L5), pupa (P) till the adult (A). Population studies on P. rapae have been carried out extensively in the countries where this species behaves as pest of the cultivated crucifers; England (Moss 1933, Richards 1940, Demspter 1967, 1968), Canada (Harcourt 1966), the United States (Pimentel 1961, Parker 1970) and Japan (Miyashita et al. 1956, Osada and Ito 1974, Ito et al. 1975, Hasui 1977). Osada and Ito (op. cit.) summarized these results: Mortality of P. rapae is commonly high at three developmental stages, Lv L5 and pupa, caused by the adverse weather (rainfall and wind), predators (paper wasps and birds) and hymenopteran parasitoids. At the area studied in the present study the amount of cultivated crucifers is so negligible that Pre highly depends upon the other larval food plants, particularly Rorippa sylvestris as clarified by adult chase for the females (Table 13). Pnn is also highly dependent upon R. sylvestris, though some amount of Oardamine leucantha, an original larval food plant, was found at the northern part of UC. As R. sylvestris received the majority of eggs laid by both species except Pre-G2, changes of the numbers of immatures on this plant could explain

Oompari8ion of Population Dynamie8 of Two Pierd Butterflies 223

Table 13. Number of eggs laid by individually marked females on various larval food plants. Rs: Rorippa sylve8triB. As: Armoraeia sustieana. So: SiBymbrium offieinale. Ra: Raphanus sativus. Be: Brassiea eampestri8. Rr: Raphanus raphaniBtrum. Ri: Rorippa iBlandiea.

-Plants received eggs

Code number of female

Rs Ar So Ra Bc Rr Ri

a;:-Pr-;: :;g bi I ~~---I--2---,-------+----·---------:----

10 I 5 13 198 15 21 21 11 23 106 22 1 20 143

Total (%) 446 (98. 5)

Pnn, '76 7 64 9 221

'77 1 18 Total (%) 303 (96.4)

G1 Pre '76 3 124 5 267

10 66 17 25 22 15 23 . 12

Total (%) 509 (66. 3) I

!Pnn '78 31 27 (96. 4) I I

G2 /Pre '7626

1

50 41 161 47 446

I Total (%)

i 657 (93. 9) i

5

7 (1. 5)

11

11 (3, 6)

2 4

2 63

2 113 9 (0. 8) 180(23.4)

1 ( 3.6)

2 3 2

4 3(0.4) 8 (1. 1)

8

8 (1. 0)

24 22

12 58 (7. 5)

1 2(0.3)

2 1

I I

5 I 8 (1. 0)

I

10 I 4

I

14 2

14(2.0) I

16 (2. 3)

the most, if not all, aspects of fluctuation of the number of adults. For this purpose, 14 quadrats (1 X 1 m) in spring, 16 in early summer and 15 in late summer were set at an area around the University buildings not so highly frequented or disturbed by men as the crop fields. In 1974 within these quadrats, 25 smaller quadrats (20 X 20 cm) were set and the numbers of individuals found therein were counted everyday, each stage separately. A needle stick was used to seek the individuals attaching on the underside of the leaves. Head width of both species was used to distinguish the instars:

Instar Ll L2 La L. L5 Head width in mm. 0.5 0.5-1.0 1.0 1.5 2.0

224 M. Yamamoto

Eggs, larvae and pupae are very alike between species, but can be distinguished by the following characters:

Color Shape

Prc Pnn Prc Pnn

Egg I creamy !

I i rounder at the I cephalic pole

thinner at the whity

Larva Pupa

I

yellowish green without dark

i patches

dark green I rod-like with dark patches stumpy

cephalic pole slightly larger flatter slender

I

Other auxiliary characters are summarized, together with some ecological notes:

Prc Egg (E) Laid singly on the underside

of leaves No particular preference on

positions of food plant Ll Often migrating to younger

leaves near the point of hatching

First lighter green, later yel­lowish

Solitary L2 Often found on the uppersides

of leaves La Mostly found on the uppersides

of leaves L4 " L5 After completion of feeding,

leaving the food plant for pupating site

Pupa(P) Pupation mostly at shaded place near food plant, especially on walls of buildings in the present study

Pnn As in Prc but often several eggs laid

jointly on the upperside of leaves With preference for shaded parts, i.e.

lower leaves or positions Very sedentary, remaining on the leaves

where laid

First darker green, later whitish

Often gregarious when oviposited jointly Not found on the uppersides of leaves

" 11

Not emigrating in post feeding period

Pupation mostly on stalks of food plant.

As pupae of Prc were mostly found on the walls of the buildings nearby, not within the quadrats, the number of pupae found on the walls was counted everyday, and the decrease of individual number in the stage later than post feeding period (PPP) was estimated based upon this census.

4.1. Decrease of individual number after PPP stage: L5/Prc was studied by releasing larvae marked with color ink at three distances, 1 m (balck), 2 m (red) and 3 m (blue) from the definite point of the wall. The migration was clearly orientated (Table 14), which caused increase of individual number invading into a quadrat as described later (p. 227). The survival rate of post-migration individuals

Marked with

Comparison of Population Dynamics of Two Pierid Butterflies

Table 14. Migration to pupating sites by marked 5th instar larvae of P. rapae crucivora. *=releasing

Black Red Blue

Found 1 I ' I No. ndi-I ! 1 I I No. indio I ! on , at 2m 3m I viduals walllim : 2m 3m viduals wall I 1m 2m 3m

Date

IX 16 17 18 19 20 21 22 23 24 25 26 27

I wall i 1m newly i ' I' newly I marked I' I I marked ,

I

I !

I

, I

I I

I ,

* 12 I * I 12 , * I I I 1 4 1 ! 12 I 9

I 11 Ij 5 7 16 2 ' 8 17

9 11 5 2 112 1 6 I

6 3 4 1 12 4 3 10 8 5 1 8

I 6 5 7 6

5 4 1 8 5 1 5 5 5 3 2

I

6 2 1 3 1 1 3 4 2 2, 2 2 , 1 1 3 2 1

I I \

1 3 2 1 4 1 4 2

1 I I I I I I 1 I

Table 15. The number of individuals of P. rapae crucivora found dead on the building walls at various stages.

Stage 'I

Site censused ,

Gt Aa Ab Ac Bd Be

Total

G2 Aa Ab Ac Bd Be

Total

Gs Aa Ab Ac Bd Be

Total I

ppp pp

I

0 ! 2 3 4 1 1 5 0 2 1

11 8 I

5 1 66 14 17 8

128 25 18 1

234 49

22 1 28 19 11 10 6 6 2 1

69 37

p

I 1 7 5 1 2

16

0 10 10 20 5

45

0 19 15 9 4

47

I Unsuccessful I No. emergence I emergence

0 0 1 4 0 2 0 1 0 0 1 7

0 0 0 0 0 1 0 3 0 0 0 4

0 3 1 6 0 0 0 2 0 0 1 11

225

No. indio viduals newly

marked

12 10 17 6 4 7

2

3

226 M. Yamamoto

can be estimated by the census on the walls at each surveyed section (Ta8le 15). Considering total number of individuals found dead as the number of individuals survived from PPP to adults (A) is given (PP=pre-pupa, P=pupa, unsuccessful emergence in parentheses):

PPP G1 43 G2 377 Ga 165

PP 32

143 96

P 24 49 59

A 7(1) 4(0)

10(1)

A/pPP 0.1628 0.0106 0.0606

The survival rate emergence/PPP was thus 16.3% (G1 ), 1.06% (G2) and 6.06% (Gs). In Pnn intense collection of pupae and parasite cocoons was made only for G2,

1979, to reinforce the insufficient result in G2 of 1974 (Table 16). The survival rate adults/pupae was 49.2% (unsuccessful emergence was 25.0%), and total survival rate emergence/pupae was 36.9%, both higher than in Pre.

Table 16. The number of G2 dead individuals of P. napi nesis collected in 1979.

No. emergence Stage I PPP I P

Site censused

Unsuccessful emergence

Ab I I 15 8 1 5 Ab II , 26 2 1 1 AblII 32 5 2 9 Ac IV 9 9 1 3

Boe 1 0 0 1 Bon 1 6 3 5 Boo 0 3 0 0

Total 84 I 33 8 24 -- ---

4.2. Decrease of individual number till the last instar: Seasonal fluctuation of the individual number of two species is given in Figs. 10 (Pre) and 11 (Pnn) for each developmental stage. Three generations of both species are clearly segregated. A higher abundance of Pre than Pnn is found in all immature stages. In these figures, however, difference of duration among stages is not considered. As the figures show the fluctuation of the number of individuals X days, a longer lasting stage obviously causes a higher total number than a real number by accumulation. To correct this the individual number must be divided by the duration of the respective stage (Table 17). The durations obtained in the field conditions do not much deviate from those obtained in the laboratory conditions (Fig. 12). The number of calculated individuals is that expected at the middle of respective stage (Ito and Murai 1977), assuming a uniform intra-stage mortality rate. But this assumption cannot stand for some stages (Figs. 13 and 14) in which the decrease, i.e. disappearance, was higher in the earlier period of the stage. This is particularly conspicuous in Ll where disappearance was highest in the first day, i.e. just

Comparison of Population Dynamics of Two Pierid Buttrefiies 227

after hatching from eggs. In such a case the corrected number of individuals X days overestimates the real number at the middle of the stage. Using the average survival duration of each stage (Table 18), the number of individuals entering the concerned stage is estimated (Table 19), but disappearance and migration of larvae make the exact estimation difficult. Disappearance of Ll and some L2 is caused by both their death and overlook, whereas increase of individual number in the later instars (L4 and Ls) is caused by immigration from other feeding places (Section 4.1.). Only the numbers of eggs and L3 can be regarded as reasonable approximation to the real number of individuals survived within the quadrats. Considering both disappearance of Ll and immigration of the older larvae, Table 20 was prepared by the records of particular individuals entering the stage concerned, together with the result in the Section 4.1. The total survival rate until emergence was estimated by multiplying the survival rate from egg stage to L5 by that from PPP to emergence:

100

Egg

200

100

.~.

.'

.. '.

,0

• ,p q,~o II

°,°000 II 0°'1°061,

O~~ __ ~_N~'~ __ ~ _______ ~r-~'~ ____ ,-,~' __ ~~~~'~ __ ~ .~ ..

May Jun J u I Aug Sep Oct

Fig. 10. Fluctuation of individual number of P. :rapae crucivora in each developmental stage, Egg, LlO L., L., L. and L5 •

228

~ ., .D E " z

:----~Q,..,...,~oo~m-mn 50 LII

I----'--.. --t~--.. Egg

M. Yamamoto

." ....

100 .

• ,p

. '. Oct

Fig. 11. Flucutation of individual number of P. napi nesis in each developmental stage.

Table 17. Duration of each developmental stage in days passed until the next stage in field conditions (asterisked figures in laboratory conditions).

Generation 1 Gl G2 Gs

Stage I N I

x S.D·I Tota.1 I N I duratIOn !

x S.D. I Tot~1 I N I duratIOn x S D I Total

. '. duration I

8.54±1. 24 I 1

52 I

4. 92±1.06I Pre Egg 1 26

1

81 4. 01±1. 05 Ll 7 3. 29±1. 89[ 11. 83 32

1

3. 17±0.88 I 8.09 44 3.07±0.70 7.08 L2 5 2.80±0.84 14.63 35 2. 34±0. 97 10.43 40 2. 53±0. 85 9.61 La 19 3. 42±1. 26 18.05 35 2. 71±1.30 13.14

1

35 2. 60±0. 81 12.21 L4 16 3. 63±1. 09 21. 68 33 2. 79±0.82 15.93 24 4. 04±1. 04 16.25 L5 7 5. 86±1. 35 27.54 19 20. 14 22 4.41±1.26 20.66

pp 4 1.0 28.54 0 1-P 6 9. 17±1.47 37.71 3 11. 33±2. 08 31. 47

Pnn Egg 22 7. 95±1. 76 48 4.98±1.12 49 4. 06±1. 01 Ll 23 5.61±1.73 13.56 23 3.30±0.88 8.28 29 3. 31±1. 07 7.38 L2 21 5. 00±1. 45 18.56 35 2. 46±1. 01, 10.74 16 3. 56±0. 73 10.93 La 21 4. 52±1. 44 23.08 19 2.42±0.84 1 13. 16 18 3.39±0.98 14.32 L4 15 3. 53±0. 92 26.61 19 2. 79±0. 71 15.95 ! 19 4. 26±0. 99 18.58 Ls 8 5. 38±0. 74 31. 99 14. 3. 64±1. 08 19.59 ! 12 4. 92±1.38 23.50 pp 11 1. 27±0. 47 33.26 38*1 0.71±0.46 20.30 i 32* 1. 34±0. 70 24.84

P 9 12. 56±1. 59 45.82 42* 7.38±0.73 27.68 1

11* 228. 62±6. 86 253.46 I

Oomparison of Population Dynamics of Two Pierid Butterflies

c o '" ~ o

5

5

GI

P. ropoe crucivora

P. napi nesis

,. " ;-....... ~ .... --.

Gz

Larval instar

,e I

I

1",-.-----.1

I .... ,-.--...

Fig. 12. Relation of stage duration in days in field and laboratory conditions.

" en c 1;;

-5 c

"

" .0 E ~ z

D Observed to the next insfar on stage

_ Disappeared

Duration in days

229

Fig. 13. Daily change of individual number of P. rapae crucivora within a particular stage.

230 M. Yamamoto

o Observed to the next instor on stage

_ Disappeared

GI G2 G3 10

0

10 u> 0 0 '" "0

'" 10 '6 ,5 0

'0 10

a; 0 .0

~ 30 Z

20

10

0

20

10

Duration in days

Fig. 14. Daily change of individual number of P. napi nesis within a particular stage.

Table 18. Apparent average duration in days passed until transfer to the next stage or disappearance or death within each stage

by 95% fiducial limit.

Generation i G1 G2 Ga

Stage N it S.E. N it S.E. N it S.E.

Pre Egg 57 5. 95±0. 85 74 4. 15±0. 37 98 3.65±0.28 Ll 37 1. 73±0. 42 81 2. 12±0. 26 86 2. 37±0. 23 L2 9 2.1l±0.81 53 2. 08±0. 26 52 2. 31±0. 26 La 22 3. 23±0. 61 50 2. 42±0. 36 50 2.30±0.25 L, 24 3.50±0.45 47 2.60±0.85 52 2. 96±0. 40 L. 16 3. 63±1. 20 38 2. 68±1. 77 40 3. 03±0. 64

Pnn Egg 35 6.1l±1. 07 76 4. 21±0. 39 81 3. 12±0. 35 Ll 34 4.53±0.81 73 2. 21±0. 29 60 2. 35±0. 34 L2 23 4.87±0.66 39 2. 36±0. 34 19 3.42±0.43 La 21 4.52±0.66 25 2. 08±0. 27 25 2.08±0.46 L, 16 3.43±0.51 28 2.46±0.36 29 3.52±0.51 L. 16

I 5.00±0.55 23 2. 74±0. 63 22 3. 64±0. 08

------ - - ---------------- -----

Oomparison of Population Dynamics of Two Pierid Butterflies 231

Table 19. The individual number in each stage with the range of 95% fiducial limit in parentheses.

Generation I Gl G. Gs

Stage Pre Egg 285 (249-332) 969 (890-1, 065) 821 (763-887)

Ll 65( 53- 86) 275(246- 314) 404 (369-447) L2 43( 32- 71) 219 (195- 250) 219 (197-246) La 37 ( 31- 45) 264(230- 310) 207 (186-232) L. 36( 32- 41) 463(422- 512) 301 (264-348) L6 42( 31- 62) 590(485- 753) 357 (295-453)

Pnn Egg 84( 71-102) 462(424- 509) 325 (292-366)

Ll I 33( 28- 40) 185 (163- 213) 100( 87-117)

L2 I 22( 20- 26) 111 ( 97- 130) 5B( 50- 65)

LSI 26( 23- 30) 161 (143- 185) 48( 42- 56)

L. 36( 31- 42) 218 (190- 255) 56( 54- 60) L6 40( 36- 45) 280(228- 363) 65( 54- 84)

Table 20. Life tables of two species. x: Developmental stage. dx : Individual number dead within the stage. 100qx: Mortality rate in the stage concerned. Asterisked figures: Corrected based upon the data in 1979.

Generation Gs

x .---~~-.

I I Pre Egg 244 184 75.4 891 483 54.2 674 308 45.7 Ll 60 25 41. 7 408 141 34.6 366 141 38.5 L2 35 9 25.7 267 25 9.4 225 33 14.7 La 26 4 15.4 242 30 12.4 192 30 15.6 L. 22 6 27.3 212 24 11. 3 162 36 22.2 L. 16 7 43.8 188 164 87.2 126 81 64.3 P 9 6 66.7 24 22 91. 7 45 37 82.2

Adult 3 2 8

Total 98.9 i 99.8 98.9

Pnn Egg 64 17 26.6 455 215 47.3 230 114 49.6 Ll 47 9 19.1 240 63 26.3 116 47 40.5 L2 38 2 5.3 177 24 13.6 69 9 13.0 Ls 36 3 5.3 153 11 7.2 60 11 18.3 L. 33 2 6.1 142 25 17.6 49 18 36.7 L. 31 17 54.8 117 113 96.6 31 28 90.3 P 14 5 35.7 4 3* 25.0* 3 3 100.0

Adult 9 1* 0

Total 85.9 I 99.7* 100.0

232 M. Yamamoto

Prc G1 1.07% G2 0.22 G. 1.13

Consequently the number of emerged adults is:

Prc 2.6 (=c3) 1. 99(=c2) 7.68(=c8)

*=observed number

Pnn 14.06% 0.32 0.0

Pnn 9*

1.48(=cl)**

**=corrected by pupal survival rate obtained from 1979's result

Mortality rate is shown in Table 20 as 100 qx. Death number, d x , is equal to the decrease of the number from one stage to the next. Stage-specific mortality rate can be estimated by counting the number of individuals dead within the stage, too (Table 21). Two mortality rates obtained from these different calculations are similar for each other, which favors the validity of the life tables obtained. In Table 20, deaths of Ll within a day after hatching are entered d x of E.

Table 21. Mortality rate (M) in each developmental stage of two species, obtained from individual records. D: Number of individuals dead. S: Number of individuals molted to the next stage.

-

I

Prc Egg 31 26 54.4 22 52 29.7 17 81 17.3 Unfound after hatching 30 31. 6 20 14.4 14 12.6

Ll 27 38 41. 5 38 81 21. 9 45 52 46.4 L. 12 24 33.3 10 69

I

12.7 9 53 14.5 L. 7 27 20.6 7 49 12.5 8 43 15.7 L. 12 23 34.3 16 35 21. 4 19 27 41. 3

Pnn Egg 15 22 37.1 28 48 36.8 32 49 39.5 Unfound after hatching 5 11. 6 16 14.7 17 21. 5

Ll 9 29 23.7 37 56 39.8 29 33 46.8 L2 2 24 8.7 9 45 16.7 12 29 29.3 L. 2 22 8.3 7 43 14.0 8 30 21. 1 L. 1 21 4.5 12 39 23.5 12 21 I 36.4 L. 17 14 54.8 42 4

I

91. 4

I

p 5 I

9 35.7 4 0 100.0

Three factors, the number of females (Section 1.3.), the number of eggs laid (Section 3.2.) and totally survival rate obtained from these life tables determine the fluctuation of adult numbers of the next generation observed on UC in 1974, that is, Pre > Pnn between species and G1 >G2 in Pre and G1 <G2 in Pnn. Only mean values of each factor were multiplied as follows, for example, G1-Prc estimation was made from 1,280 X 250 X 0.0107:

Oomparison of Population Dynamics of Two Pierid Butterflies

In Pre

In Pnn

Observed (0) 2,539 1,844

Number of adults Estimated (E)

3,424 1,376**

G1 1,834 2,091 G2 2,030 538

D* 34.9 25.4

14.0 73.5

*=% ratio of deviation 10-EI/0 **=corrected by 66.3% dependence on Rorippa sylvestris

1,000

500

1,000

500

~

0

~ .0

E p " A z 1,000

, ,.

10 20 30 40 Wintering

Days since egg laying

Fig. 15. Survivorship curves for three generations of P. rapae erucivora and P. napi nesis.

233

234 M. Yamamoto

Except Pnn-G2 observed and estimated figures don't deviate so much for each other, judging from the correlation coefficient between adult numbers obtained by census and capture-recapture method. The high deviation in Pnn-G2 was probably caused by the overlook of some pupae which are more difficult to discover than Pre by their habit of pupating in the foliage of food plant (p. 224). Crowded weeds in summer season probably accelerate difficulty for discovering the pupae.

From the obtained survivorship curves (Fig. 15), those of female adults (Suzuki 1978) and age-specific birth rate (Fig. 7), the net reproductive rate per generation, RO) generation time, T, and intrinsic rate of natural increase, r, were calculated between generations and between species as follows (for example of Prc­G2 in Table 22):

Pre Pnn Ro T r Ro T r

G1 3.709 42.19 0.0311 22.038 54.21 0.0571 G2 0·764 34.53 -0.0078 0.505 40.76 -0.0168

where r = J!l:o = In L.TZ"m" . r would be higher III Pre than Pnn

Table 22. Daily number of survived individuals inferred from Fig. 15 and Table 20 in G2-female of P. rapae crucivora. x: Age in days. mx: Age­specific birth rate.

x Ix dx mx x

0-1 1,000 136 26-27 - 2 864 136 -28 - 3 728 135 -29 - 4 593 135 -30 - 5 458 89 -31 - 6 369 28 -32 - 7 341 26 -33 - 8 315 15 -34 - 9 300 14 -35 -10 286 14 -36 -11 272 11 -37 -12 261 11 -38 -13 250 12 -39 -14 238 14 -40 -15 224 13 -41 -16 211 18 -42 -17 193 18 -43 -18 175 18 -44 -19 157 55 -45 -20 102 75 -46 -21 27 1 -47 -22 26 1 -48 -23 25 1 -49 -24 24 1 -50 -25 23 1 -51 -26 22 14.8 -52

I -- .. ------~~ ------

Ix

7.2 6.2 5.2 4.2 3.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.2 2.0 1.7 1.7 1.7 1.7 1.7 1.4 1.2 1.0 0.8 0.3 0.2 0.2

dx

1 1 1 1 1 0 0 0 0 0 0 0 0.2 0.3 0 0 0 0 0.3 0.2 0.2 0.2 0.5 0 0.3

mx

o 20 60

125 120 90 60 50 40 35 30 30 25 25 20 10 5 o o o

under

Comparison of Population Dynamics of Two Pierid Butterflies 235

laboratory conditions due to higher fecundity of Pro (cf. Southwood et. al. 1974), while higher in Pnn than Pro under field conditions in the present study. Most previous reports estimated r in experimental conditions in which the difference of the total survival rate among compared species was ignored.

4.3. Mortality factors: In Table 20 the high mortality rate in two species appears at three developmental stages. The first high mortality occurs in the early stage of Lv which has already been recognized by the life tables prepared for the populations feeding on cultivated crucifers and is caused both by adverse weather and insect predators. The second one appears in Ls, caused partly by a parasitoid wasp, Apanteles glomeratus. The third one in pupal stage is caused partly by another parasitoid wasp, Pteromalus puparum which seems more serious in Pro than Pnn. In addition to these well-known mortality factors, other factors are shown in detail in Table 23 in which cases of deaths are individually mentioned. But d x of L1 ", L4 in Table 20 is much higher than deaths actually observed. Most cases of these disappearance seem to be caused by tabulated predators and un­known nocturnal predators (Dempster 1967). On the deaths at stages after Ls the result in Table 23 seems to reflect the real relative importance. The kinds of predators would be different according to the food plant species and localities surveyed, though a fairly high mortality is caused by them, especially at early developmental stage by insect predators (about 48% in E, 59% in Ll of Pro while 33% in Ll of Pnn) and later stage by vertebrate predators (the most cases of disappearance may be caused by them, i.e. 26% in PPP and 19% in PP of Pro, though not surveyed in Pnn).

4.4. The relation between mortality rate and population density: Nextly the relation between these mortality factors and population density is thought by key­factor analysis (Morris 1963a, Kuno 1973, Southwood and Reader 1976, Podoler and Rogers 1975, Ito and Murai op. oit.). The logarithmic number of individuals determined at a given developmental stage is plotted to that at the next stage. Then equation of the regression line and the correlation coefficient, r, are ca lculated. From the coefficient of the regression line, b, the relation between mortality rate and population density can be suggested, that is, density-dependent when b<l, density-independent when b=l and inverse density-dependent when b>l. Density-dependent mortality can regulate a population within a stable range, acting intensely at high densities but weakly at low densities (cf. Roughgarden 1971, May et al. 1974, Stubbs 1977). On the other hand, density-independent and inverse density-dependent mortality would ultimately increase the density if not accompanied with density-dependent mortality. In the present paper key factors mean those which cause density-dependent mortalities. In this method r2 is called the coefficient of determination, being used as the standard of the rate of the change of dependent variables affected by the change of independent variables. The scatter diagrams (Figs. 16",24) show the relations between egg and Ll (E-Ll)' LCL2'

236 M. Yamamoto

Table 23. The number of individuals the deaths of which were directly observed. Asterisked items were obtained not from the quadrat.survey.

Generation G1 G2 G3

Stage Mortality factors

Prc Egg " ........................ 24(%) 69(%) 54(%) Predacious mites 4( 16.7) 39( 56.5) 28 ( 51. 9) Detached from leaves 20 ( 83.3) 6( 8.7) 14( 25.9) Unfertilized 17 ( 24.6) 9 ( 16.7) Eaten by older larvae 7 ( 10.2) 3( 5.5)

Ll ., ..................... ,., 2 4 11 Spider mite 1 ( 50.0) 2 ( 50.0) 7 ( 63.6) Unsuccessful molting 1( 9.1) Disease 1 ( 25.0) Spider 1 ( 9.1) Mishandled 1 ( 25.0) 1{ 9.1) Detached from leaves 1 ( 9.1) Unknown 1 ( 50.0)

L2 .......................... 5 Unsuccessful molting 3 ( 60.0) Unknown 2 ( 40.0) 1 (100. 0)

La ••••••••••••••• , ••••.•.• o. 3 Unsuccessful molting 2( 66.7) Unknown 1 (100. 0) 1(100.0) 1 ( 33.3)

L4 .......................... 7 8 Unsuccessful molting 1 ( 14.3) 4( 50.0) Disease 1 (100. 0) Unknown 6 ( 85.7) 4( 50.0)

L. .......................... 3 46 7 Apanteles glomeratus 2( 66.7) 29( 63.0) Spider 1 ( 14.3) Dermapteran 1 ( 14.3) Unknown 1 ( 33.3) 17(37.0) 5 ( 71.4)

PPP.* .......................... 11 234 69 Apanteles glomeratus 6 ( 54.5) 159 ( 67.9) 19 ( 27.5) Disappearance 5 ( 45.5) 47 ( 20.1) 30( 43.5) Spider 4( 1. 7) 13 ( 18.8) Disease 24 ( 10.3) 7 ( 10.2)

PP*. " .••.•.•..••...•••.••. , o. 8 49 37 Disappearance 8( 16.3) 10(27.0) Unsusccessful pupation 8 (100.0) 41 ( 83.7) 27( 73.0)

P.* .......................... 16 44 47 Pteromalus puparum 10 ( 62.5) 14( 31. 8) 11 ( 23.4) Other parasitoid waps 2( 12.5) 1 ( 2.3) 5( 10.6) Disappearance 4( 25.0) 23 ( 52.3) 29 ( 61. 7) Disease 5 ( 11.4) Drying 1 ( 2.2) 2( 4.3)

P. .......................... 5 Pteromalus puparum 2( 40.0) Artifacts 3( 60.0)

Oomparison of Population Dynamics of Two Pierid Butterflies 237

Table 23. (Continued)

Generation

Stage Mortality factors I

Emergence* . ...... ............ ....... , 1(%) 1(%) Unsuccessful emergence 1 (100. 0) Wing undeveloped 1 (100. 0)

Adult. * .......................... 2 2(%) 1 Adverse weather 1 ( 50.0) Sparrow 1 ( 50.0) Traffic accident 1 (100. 0) Artifacts 1 ( 50.0) Spider 1 ( 50.0)

Pnn Egg .......................... 1 14 25 Predacious mites 13( 52.0) Detached from leaves 3 ( 21. 4) 5 ( 20.0) Unfertilized

I

1 (100. 0) 11 ( 78.6) 6 ( 24.0) Eaten by older larvae l( 4.0)

Ll .......................... 5 1 Spider mite 2 ( 40.0) Mishandled 1 (100. 0) Unknown 3( 60.0)

L. .......................... - - -I

La .......................... 1 1 1 Unsuccessful molting 1 (100. 0) Spider 1 (100. 0) Unknown 1 (100. 0)

L. ......................... 10 2 Unsuccessful molting 2( 20.0) Apanteles glomeratus 3 ( 30.0) Carabid bettIe l( 50.0) Unknown 5 ( 50.0) 1 ( 50.0)

Ls .......................... 46 1 Apanteles glomeratus

, 42( 91.3) 1 (100. 0)

Unknown i

4( 8.7)

PPP .......................... 84 Apanteles glomeratus 77( 91. 7) Disease 7( 8.3)

P*. .......................... 33 Pteromalus puparum 20( 60.0) Other parasitoid wasps 5 ( 15.2) Disease or drying 6 ( 18.2) Diptera (Tachinidae) 2( 6.0)

P. .......................... 4 1 Pteromalus puparum 2( 50.0) Other parasitoid wasps 2( 50.0) 1 (100. 0)

Emergence. * .......................... 8 Wing undeveloped 8(100.0)

238 M. Yamamoto

L2-La, Ls-L4' L4-L5, Ls and pre-pupa (L5-PP), prepupa and pupa (PP-P) and pupa and adult (P-A). The mortality factors for each stage are inferred as follows:

Stage Symptom Principal mortality factors Generations* Prc Pnn

E-Ll Disappearance Detached and fallen from leaves 1 by adverse weather (not dis-tinguished from Ll deaths just after hatching)

Darkened Unfertilized 2 1,2 Shrivelled Sucked by spider mites 2,3 3 Leaf, to which E Eaten by older larva with leaf

attached, eaten Ll-L. Dead without molting Unsuccessful molting

Disappearance Predators (mainly spider mites 1,2,3 and spiders)

Darkened Disease, etc. L.-L. Disappearance Predators (insects and birds) 3 1,3

Dead without molting Unsuccessful molting 2,3 2 Changing yellowishly Disease 1

L.-PP Accompanied with Apantele8 (flomeratUB 1,2 2 parasitoid coccoons

Disappearance Predators (birds and spiders) 3 PP-P Disappearance Predators (birds)

Dead without pupa- Unsuccessful pupation 1,2,3 tion

P-A Disappearance Predators (birds) 2,3 With small emergence PteromlaUB puparum 1 2

holes With large emergence Hymenoptera (Ichneumonidae)

holes and Diptera (Tachinidae) Darkened Disease Dried up Unknown factor

A Dead without emergence, though Unsuccessful emergence 1,3 with pigmented wing

*=generations affected more than 50%

Although the influence of these mortality factors cannot be expressed quantitatively yet, some comments are given on the mode of action of these factors based upon the values of b at particular stages of each generations.

Pre:

E-Ll (Fig. 16): In the course of the census deaths of Ll were frequently observed within a day after hatching. Possibly deaths in Ll are likely to appear as a lasting influence of mortality factors during egg stage. Density-dependent rela­tion in G1 and G2 is possibly caused by concentration of predators, especially in spring in the areas of high egg densities, which are usually well insolated.

"0 ., i':': ~ .0 o

Q) a > E ~

a u; .S

10

b ~ ., .0 E '" z

2

2r

• P. rapae crucivora P. napi nesis

GI

y= O. 737x - 0.242 0

r2=0.886

G2 y=0.866)(-0.083 0/ r2=0.853

GI y=0.949x-0.10

r'=0.960

G2 y= 0.910x -0.132

r2=0.905

G3 . /

y=I.016x - 0.297 / I G3

r2=0.920 0 /{ "'.'"

/ ..

./

.. /

OLI---------r--------~-----~--------,_--------_r----2 o 2

Number of eggs observed

Fig. 16. Relation of the number of eggs and 1st instar larvae (L1).

-g

~ .0 o

'" o > .l'i

.~ "0 C

N

"0

'" .0

~ Z

P. rapae crucivora P. nap; nesis GI GI y=1.012x - 0.269 Y =0.942x - 0.039.

r2=0.886 r2=0.968

6---6_ ,LI LL _________ -----~ 2r

2

o

G2

y=0.975x-0.156 / Gz y=I.050x - 0.210

r2=0.948 r2=O 967 'Yo

/ 0/0 /. .

G3 /1 G3 y=0.995x-0.204 /' y=1.020x-0.Z72

2_ /'

r _0.916/0 00"

00

00 o

,2=0.667

.'.

./

,~.

~o L-~/ ____ ~ ____ ~ 2 2

Number of 1st instar larvae observed

Fig. 17. Relation L1-L!.

~ ~ ~ ~.

~ .sa.. '"'ci ~ f 6' O!

f ~.

.sa..

~ '"'ci [

~ ~ ~.

~ <C

"C

" >

" ~ .0 o

" ~ .2

c

~ 1:

'" "0 :;; .0

§ Z

P. ropae crucivora P. nap! nesis

GI GI ~ yOI.o.7o.x-o..~ . 70

0..954

o 0 1. / " ,

Y oo..832x + 0..0.

,200..9660

G2 /G2 :/ y ol.o.o.4x-o..o.61 0 0 yoo..944x - 0..0.13 •

r20 o..965 ,200..982

"Y •• d" /,.

2 G3 yoo..945x + 0..0.0.2 /

"~?

./ /

G3

Y°1.153x - 0..20.1 /

,200..832 /

;I

o.V '~.'----r----~ 2 2

Number of 2 nd instar larvae observed

Fig. 18. Relation L2-L3 .

"C

" 1: ~ .0 o

" c 1: .2

.~

.c

'" "0 :;; .0

§ Z

P. ropae crucivoro

GI

yoo..894x- ?O ,200..830. 0

P. napi nesis

GI

yoo..987x - 0.

,200..984

7

./

z::---o

0/ Gz G2 yol.o.39x - 0..112 YOl.o.45X-o../ ,200..956 0/ ,2"0..972

/ / .. o 0 0 0 0 /. ..

2-r -0 ,~

G3 / y o 0..953 x - 0.0.19

G3

,2~ 0..987 ",0

yo" 0/00

yoo..884x - 0..0.12

,200..842 /

O~ / , 2

Number of 3 rd ins tar larvae observed

Fig. 19. Relation L.-L."

*

~ ~ ;§ ~ ;§ c ;SO

'0 Q)

i': :l: .D o

~ i': B

~ .S'

'" I!)

b Q)

.D E ::0 Z

P. rapae crucivoro P. napi nesis

GI GI /

y~0.795x-0.012 Y~1.03IX-0'07

'c,"" 7 2 ~-------~--------.

G2 /G2/ yo 1.077x - 0.163 0 Y ~1.012 x - 0.130 •

r200.939 / r 20 0.962 ¥ •

,/ / o / L .~

2- '-G3

Y~0.934X-0'05y G3 r20 0 -

.942 0 Y - 0.791 x - 0 0 . 56

<f , ,,';(- """/

,,/ /" 2 2

Number of 4 th instar larvae observed

Fig. 20. Relation L.-L5'

'0

~ :l: .D o Q)

" a. ::0 a. ~ a.

b

~ .D E ::0 Z

GI y~0.898x-0.027 0

r2~0.871

o o

~ 0 G2 y~0.600x - 0.112

r2~ 0.575 o

o

o 00/ 0 o 0

Q 0/ 0

0---o-o--0--r0-Q-----~

~:0.793X-0.067g 0 /

r2~0.788 0 /

o 0/00 0;/ /0000

00---0--0----,--------

Number of 51h inslar larvae observed

Fig. 21. Relation L5-PP in P. rapae crucivora.

f ~o

.Q., ~ ~ ~ gO

f ~.

.Q.,

t ~ & . ... ~

~ ~ &' '"

t:l >I'­.....

"0

'" c: :l: .0 o

'" " c. " Q.

o ~

'" .0 E " z

GI /

Y=0.93~:+0.~ ,2=0.99/

oV/ G2

y=0.692x - 0.035 0

,2=0.710

o /0

o

o

o

/0 Ob~--o-o--o'---r-----------

G3 o y=0.763x- 0.019 0

,2=0.719 o

o

0/00

06-' -0--,-0--.,..---------

Number of pre-pupae observed

Fig. 22. Relation pp.p in P. rapae crucivora.

"0 Q)

~ .g ., o

" Q)

E' ., E ., 'l5

'" .0 E ~

GI y=0.276x - 0.007

~ 6=--0-0 I

GI y=0.427x' 0.177

"oo~

0--0 i

G2 Y = 0.562x - 0.04%

r2=0.3!lo4

00-/-----0--0'----,-------

Number of pupae observed

Fig. 23. RelatIOn P·A in P. rapae crucivora.

"0

~ ~ .0 0 ., " g-o.

'0 ~ .c § Z

0

"0

'" E' '" E '" j!! "S "0

" 'l5

1l E 2 00

/ .

-GI

-G2

.----.. -.---.,----------~ Number of 5 th instar larvae observed

----, 2

Number of pupae observed

Fig. 24. Relation L •. P and P·A in P. napi nesis. G1 from two points.

~ II>­~

~

r

Oomparison of Population Dynamics of Two Pierd Butterflies 243

LI-L2 (Fig. 17): Mortality rate is high, but seemingly governed by density­independent factors. The migration of LI to adjacent leaves may cause randomness of prey during this stage.

L2-L3 (Fig. 18), L3-L4 (Fig. 19) and L4-LS (Fgi. 20): Mortality factors in these stages seem to act density-independently except in Gv L4-LS at which density-dependent influence is suggested.

Ls-PP (Fig. 21): Highly density-dependent except GI due to the high in­cidence of attack by Apanteles glomeratus which lays eggs in L2 (Sato 1976).

PP-P (Fig. 22): Density-dependent stage except GI , mainly by unsuccessful pupation, being caused by lack of food source at late larval stage.

P-A (Fig. 23): Highly density-dependent by the parasitoid wasp, Ptero­malus puparum, and predation by birds which tend to forage continuously at favorable sites, the latter being particularly serious by immobility of pupae.

A- : Unsuccessful emergence is assumed to be caused by adverse weather (density-independent), but reliable data are not yet sufficient.

Pnn:

E-Ll (Fig. 16), LI-L2 (Fig. 17), L2-L3 (Fig. 18), L3-L4 (Fig. 19) and L4-L5 (Fig. 20): Density-independent in contrast to Pre because of sedentarily remain­ing on the underside of the leaves until L4•

L5-P (Fig. 24): By the absence of migration, this stage is not subdivided in two stages as in Pre. Inverse density dependence in GI suggested by the value of

Table 24. Regression equations between two developmental stages.

Generation I

Stage Prc E-Ll y=O.7371x-O.2417 y=O.8663x-O.0843 y=1.0156x-O.2971

LI-L2 y=l.01l7x-O.2691 y=O.9745x-O.1562 y=O.9947x-O.2040 L2-L. y=O.8318x+O.0668 y=l.OO44x-O.061O y=O.9448x+O.OO22 L.-L. y=O.8936x-O.0534 y=l.0390x-O.11l8 y=O.9534x-O.0187 L.-L5 y=O.7952x-O.01l7 y=l.0773x-O.1625 y=O·9338x-O.0521

L5-PP I y=O.8980x-O.0268 y=O.6002x-O.l116 y=O.7927x-O.0672 PP-P y=O.9348x+O.OO50 y=O.6922x-O.0346 y=O.7631x-O.0192

P-A I y=O.2760x-O.OO68 y=O.4271x-O.1766 y=O.5618x-O.0426

Total y=O.1071x-O.OO73 I y=O.1874x-O.3674 y=O.2995x-O.2655

Pnn E-Ll y=O.949lx-O.l034 y=O.9104x-O.1324 y= 1.1237x-O.4816 LCL2 y=O.9415x-O.0387 y=l.0496x-O.2104 y=1.0l95x-O.2716 L.-L. y=l.0700x-O.l082 y=O.9437x-O.0125 y=1.1530x-O.2009 L.-L. y=O.9869x-O.0366 y=l.045lx-O.0908 y=O.8841x-O.0123 L.-L. y=1.0305x-O.0852 y=l.0121x-O.1297 y=O.7913x-O.0563 L.-P y= 1.2263x-O.5681 y*=O.7952x-O.l119

i

-P-A y=1.3672x-O.5220 y*=O.7662x-O.0613 -

Total y=1.6314x-1.9377 y*=O.5685x-O.4818 I

I ----

*=Corrected based upon the data in 1979

244 M. Yamamoto

b is possibly caused by gregarious habit. G2 is density-dependence as in Pre due to the parasitism by Apanteles glomeratus.

P-A (Fig. 24): A high value of bin G1 is not so reliable as calculated from the data of only two quadrats. A density-dependent tendency in G2 is probably

'" "3

" o

'" .0 E '" z

P. rapae crucivora

P. napi nesis

o

2

Number of eggs

Fig. 25. Relation E-A. G2*: Corrected based upon the data in 1979.

caused by depredation would not so high as Pre by the pupal sites covered with snow in this species.

Total regression equations between E-A were estimated as given in the last column of Table 24. A higher density-dependent mortality in Pre than in Pnn is also shown in Fig. 25. In summarization, key factors are parasitism by Pteromalus puparum and predation by birds at P-A stage both in Pre and Pnn.

Concluding remarks

Two related pierid species feed on a common food plant, Rorippa sylvestris on the University Campus. This plant was introduced to Hokkaido from Europe

Comparison of Population Dynamics of Two Pierid Butterflies 245

before several decades (Ito, personal communication). Previously Pnn has mainly fed on Cardamine leucantha in woodlands, while Pre on cultivated crucifers. Such populations feeding on Cardamine in shaded areas are still kept in woodlands near Sapporo (Jozaneki and Hakkenzan). Since the proliferation of R. sylvestris, both species would begin to adopt this exotic plant as larval food. Its dense foliage would have allowed Pnn to live in open area, though females of Pnn do not lay eggs on the poor foliage of Rorippa at highly insolated areas. This makes habitat segregation between two species obscure, especially in adults, though Pre still prefers open area and Pnn shaQ.ed areas. For these pierid species inhabiting UC, R. sylvestris has an overwhelming importance than other crucifers, cabbages, radishes, etc., by its stability as a food source. It is perennial, abundantly grows in high density and constantly bears fresh and soft leaves preferred by young larvae. Different adult abundance between species and among generations, i.e. Gh>G1 >G2 in Pre Gh<G1 <G2 in Pnn, Prc~Pnn in Gh, Prc>Pnn in G1 and Pre <Pnn in G2 was probably maintained in 1974 by following main factors:

Gh········ .... Pre ~ Pnn ......... , .. unknown V II

G1 ••·•·•·•·· .. Pre > Pnn ......... , .. different fecundity V II

G.·········· .. Pre < Pnn ............ different mortality rate

different mortality different mortality rate between Gh rate and GH but different fecundity

between G1 and G.

Further, these two pierid species are different in the following aspects as elucidated in the present study:

Eggs Size Number

Larvae and pupae Sedentariness Density-dependence in mortality

Adults Generation time Body size r under laboratory conditions Dispersal ability of females

Pre Pnn

< >

< >

< > > >

Various features in the mode of life shown above can be interpreted by the r- and K- selection theory (MacArthur 1960, Pianka 1970, Southwood et al. 1974, Southwood 1977). The ecological strategy of a given species will be determined by

246 M. Yamamoto

the degree of the stability of its habitat, i.e. K-selection more prevailing in permanent and predictable habitat and r-selection in temporal and unpredictable habitat. But such the generalization is not always appropriate for the present case in which two related pierid species recently began to depend upon a new common food source, R. sylvestris (cf. Young and Muyshondt 1972, Sterns 1976). Although Pre may be originally r-strategist and Pnn K- strategist, their ecological status on UC is still uncertain because they have barely adopted recently the new mode of life, for Pre the adoption of perennial food plant, for Pnn the life in open­land. But a hesitating prediction could be presented on the future of two species as to their relative abundance. Density-dependent mortality, which acts more intensely in Pre than in Pnn, would cause proportional increase in mortality rate or decrease in fecundity as population increase (Stubbs op. cit.). To avoid the density-dependent mortality, the egg density should be lowered. As far as the butterflies are concerned, decrease of egg density does not necessarily require the decrease of the number of eggs laid per female. A low egg density without decreasing the number of eggs laid per female could be realized either by a high dispersal ability or by a sparse oviposition by females as actually adopted by Pre. On the other hand, Pnn shows a tendency to increase the egg density within an area by its sedentary nature and concentration of eggs. These differences of life mode will cause increase of Pnn up to a certain population density relative to that of Pre, the limit of which could be clarified only by further analysis.

Summary

Population studies on two related pierid species, Pieris rapae erueivora (Pre) and P. napi nesis (Pnn) living in the same habitat and feeding on the same food plant were carried out on the campus of Hokkaido University from 1972 to 1979. Main results are:

1. The number of adults counted by belt transect census related exponen­tially with that estimated by the caputre-recapture method, i.e. the absolute density of adult butterflies can be estimated by combining the capture-recapture method and belt transect census.

2. Dominance of Pre in earlier years followed by increase of Pnn since 1974 was inferred.

3. In 1974 abundance of adults among three generations (post-hibernating generation=Gh , 1st generation=G1 and 2nd generation=G2) was Gh >G1 >G2 in Pre while G2>Gl >Gh in Pnn.

4. The number of eggs laid per female in Pre was twice or thrice as many as in Pnn, that is, 600,..., 800 in Pre and 200,..., 400 in Pnn.

5. Duration of stay of adult butterflies within a limited area was longer in males than in females in Pre. Between two species, Pnn was more sedentary than Pre in Gcfemales.

6. Total survival rate obtained from the results of life tables was: Pre-Gl ;

Comparison of Population Dynamios of Two Pierid Butterflies 247

1.07%, -G2 ; 0.22%, -G3 ; 1.13%, Pnn-G1 ; 14.06% and -G2 ; 0.32%. 7. Female number, number of eggs laid per female and total survival rate

elucida ted a considerable part of the fluctuation of the adult number estimated on the studied area in 1974.

8. The estimated intrinsic rate of natural increase, r, was higher in Pnn than in Prc, though considerably fluctuated among generations.

9. Considering the relations between mortality rate and density, the influence of density-dependent mortality was more intense in Prc than in Pnn.

Acknowledgements

I wish to express my sincere thanks to Prof. Shoichi F. Sakagami, Institute of Low Temperature Science, Hokkaido University, for his kind guidance to the present study and Prof. Mayumi Yamada, Zoological Institute, Hokkaido University, for his reading through the manuscript. Particular appreciation is due to Mr. Takeshi Ohtani, Zoological Institute, Hokkaido University for his unselfish help at tedious field surveys of adult chase. Cordial thanks are also due to all members of the staff of the Experimental Farm, Faculty of Agriculture, Hokkaido University, especially to Prof. Fumiji Kita for his allowance of butterfly census and other activities, and to Prof. Koji Ito, Graduate School of Environmental Research, Hokkaido University, for his valuable advices on the larval food plant, Rorippa sylvestris.

References

Dempster, J.P. 1967. The control of Pieris rapae with DDT. I. The natural mortality of the young stages of Pieris. J. appl. Ecol. 4: 485-500.

---- 1968. The control of Pieris rapae with DDT. II. Survival of the young stages of Pieris after spraying. Ibid. 5: 451-475.

Frazer, D.G. 1973. Estimating butterfly numbers. Biological conservation 5: 271-276. GOda, K. 1969. The seasonal distribution of the butterflies at Yanagida, Utsunomiya.

Insecto Utsunomiya 20: 9-19. (In Japanese) Harcourt, D.G. 1966. Major factors in survival of the immature stages of Pieris rapae (L).

Canad. Ent. 98: 653-662. Hasegawa, J. 1966. Seasonal variation of fore-wing length in P. melete and P. napi nesis.

Coenonympha 18: 343-345. (In Japanese) Hasui, H. 1977. On the seasonal variability of survivorship curves and life tables of Pieris

rapae crucivora Boisduval (Lepidoptera: Pieridae). Jap. J. Ecol. 27: 75-82. (In Japanese)

Hirose, K. 1954. Weather conditions and daily activity of Pieris rapae crucivora. Shin Konchu 7: 33-39. (In Japanese)

Hiura, I. 1973. Butterflies migrate across the sea (in Japanese). iv+200 pp. Aoki, Tokyo. ----1979. An outline of genus Pieris. Acta Rhopalocerologica 3: 1-19 .(In Japanese) Ito, Y. and M. Murai 1977. On the research method for animal ecology (in Japanese). iv+

558 pp. Kokinshoin, Tokyo. ----, Sakiyama, M. aud M. Osada 1975. Population dynamics of Pieris rapae

crucivora Boisduval (Lepidoptera: Pieridae), an introduced insect pest in Okinawa. II. Features of the population dynamics based on the results of one-year survey. Jap. J. app!. Ent. Zoo!. 19: 29-34.

248 M. Yamamoto

Iwao, S. 1971. Estimstion of animal populations by caputre·recapture method. Biological Science (Tokyo) 23: 14---22. (In Japanese)

Iwata, T. 1967. Large eggs. In M. Morisita and T. Kira (eds.), Natural history­Ecological studies, Ch. VIII, p. 211-248 (in Japanese). Chuokoron·sha, Tokyo.

Kawabata, M. 1976. The species composition and the seasonal change of butterflies in the suburbs of Fukuoka - A record. Physio!. Eco!. Japan 17: 359-363. (In Japanese)

Kobayashi, S. and S. Takano 1978. Survival rate and dispersal of the adults of Pieris rapae crucivora Boisduva!. Jap. J. app!. Ent. Zoo!. 22: 250-254.

Kuno, E. 1973. Statistical characteristics of the density-dependent population fluctuation and the evaluation of density-dependence and regulation in animal populations. Res. Popu!. Eco!. 15: 99-120.

Kurata, M. 1955. A study on the egg size, egg number on a food-leaf and the larval growth with reference to determine each instar of Luehdorjia puziloi inexpecta Sheljuzhko. New Entomo!. 4: 23-26.

MacArthur, R. 1960. On the relative abundance of species. Amer. Natur. 94: 25-36. May, R.M., Conway, G.R., Hassell, M.P., and T.R.E. Southwood 1974. Time delays,

density-dependence and single-species oscillations. J. Anim. Eco!. 43: 747-770. Miyashita, K., Ito, Y. and A. Gotoh 1956. Study on the fluctuation of egg and larval

popultions of common cabbage butterfly and the factors affecting it. Studies on the mortality rate and behavior of field population of agricultural pest insect (3rd rept.). Jap. J. app!. Ent. 12: 50-55. (In Japanese)

Morisita, M. 1967. The seasonal distribution of the butterflies in the suburbs of Kyoto. In M. Morisita and T. Kira (eds.), Natural history - Ecological studies, Ch. IV, p. 94-132 (in Japanese). Chuokoron-sha, Tokyo.

Morris, R.F. 1963a. pp. 30-37 in The dynamics of epidemic spruce budworm populations. (ed. R.F. Morris), Mem. Ent. Soc. Canada 31.

Moss, J.E. 1933. The natural control of the cabbage caterpillars, Pieris spp. J. Anim. Eco!. 2: 210-231.

Ohsaki, N. 1978. In pursuit of pierid populations: Adaptive strategies on the three related pierid species. Shizen 389: 56-67. (In Japanese)

Okuyama, S. 1963. Concerning two naturalized plants. Jour. Jap. Bot. 38: 256. Osada, M. and Y. Ito 1974. Population dynamics of Pieris rapae crucivora Bosiduval

(Lepidoptera: Pieridae), an introduced insect pest in Okinawa. I. Winter-spring generations. Jap. J. app!. Ent. Zoo!. 18: 65-72.

Parker, F.D. 1970. Seasonal mortality and survival of Pieris rapae (Lepidoptera: Pieridae) in Missouri and the effect of introducing an egg parasite, Trichogramma evanescens. A. Entomol. Soc. Amer. 63: 985-994.

Pianka, E.R. 1970. On r- and K-selection. Amer. Natur. 194: 592-597. Pimentel, G. 1961. Natural control caterpillar populations on cole crops. J. Econ. Entomol.

54: 889-892.

Podoler, H. and D. Rogers 1975. A new method for the identification of key factors from life-table data. J. Anim. Eco!. 44: 85-114.

Richards O.W. 1940. The biology of the small white butterfly (Pieris rapae), with special reference to the factors controlling its abundance. J. Anim. Eco!. 9: 243-288.

Roughgarden, J. 1971. Density-dependent natural selection. Ecology 52: 453-468. Sakagami, S.F. and H. Fukuda 1973. Wild bee survey at the campus of Hokkaido

University. J. Fac. Sci. Hokkaido Univ. Ser. VI, Zoo!. 19: 190-250. Sato, Y. 1976. Experimental studies on parasitization by Apantele6 glomeratus L.

(Hymenoptera: Braconidae). I. Parasitization to different species of genus Pieris.

Oomparison of Population Dynamics of Two Pierid Butterflies 249

App!. Ent. Zoo!. 11: 165-175. Southwood, T.R.E. 1977. Habitat, the templet for ecological strategies? J. Anim.

Eco!. 46: 337-365. ---~, May, R.M., Hassel, M.P., and G.R. Conway 1974. Ecological strategies and

population parameters. Amer. Natur. 108: 791-804. --~ and P.M. Reader 1976. Population census data and key factor analysis for the

viburnum whitefly, Aleurotrachelus jelinekii (Frauenf.), on three bushes. J. Anim. Eco!. 45: 313-325.

Stearns, S.C. 1976. Life-history tactics: A review of the ideas. Q. Rev. Bio!. 51: 3-47. Stubbs, M. 1977. Density dependence in the life-cycles of animals and its importance in

K- and r-strategies. J. Anim. Eco!. 46: 677-688. Suzuki, Y. 1978. Adult longevity and reproductive potential of the small cabbage white,

Pieris rapae crucivora Boisduval (Lepidoptera: Pieridae). App!. Ent. Zoo!. 13: 312-313.

---, Nakanishi, A., Shima, H., Yata, 0., and T. Saigusa 1977. Mating behaviour of four Japanese species of the genus Pieris (Lepidoptera: Pieridae). Kontyu 45: 300-313.

Yamamoto, M. 1975. Notes on the methods of belt transect census of butterflies. J. Fac. Sci. Hokkaido Univ. Ser. VI, Zoo!. 20: 93-116_

~--- 1977. A comparison of butterfly assemblages in and near Sapporo City, northern Japan. Ibid. 20: 621-646.

~-- 1978. Number of eggs laid by Pieris napi nesis Fruhstorfer (Lepidoptera, Pieridae) in Sapporo, northern Japan. New Entomo!. 27: 61-66.

--- and T. Ohtani 1979_ Number of eggs laid by Pieris rapae crucivora, compared with P. napi nesis, in Sapporo (Lepidoptera: Pieridae). Kontyu 47: 530-539.

Young, A.M. and A. Muyshondt 1972. Biology of Morpholyphemus (Lepidoptera: Morphidae) in El Salvador. J. New York Entomo!. Soc. 80: 18-42.